counterpoise.py: Counterpoise-Corrected Binding of Two Molecules

The script counterpoise.py can be used to calculate a counterpoise-corrected binding energy of a complex consisting of two non-covalently bound molecules. You can run this script from Maestro, as described in the Counterpoise Panel help topic, or from the command line using the following syntax:

jaguar run counterpoise.py filename

where filename can be either a Jaguar input file or a Maestro structure file. A job directory called filename_counterpoise is created in the current working directory, and the job files for all calculations are written to this directory. When the job finishes, a file named filename_counterpoise.out is written to the current working directory. This file contains the counterpoise-corrected binding energy and the counterpoise correction energy.

If you use a Jaguar input file, you can make settings to customize the calculation. However, you should not mark the atoms in the input file as counterpoise atoms (see Counterpoise Calculations), as this is done automatically in the input files that are generated by the script.

If the basis set and level of theory are not specified in the input file, then each job runs using the same default settings as for Jaguar jobs launched from Maestro: the default level of theory is DFT with the B3LYP functional, the default basis set is 6-31G**, and no geometry optimization is performed. You can specify keyword settings in the input file if it is a Jaguar input file, but you can also specify settings on the command line. For example, if you want to optimize the geometries of the fragments and of the complex using the X3LYP functional, then you would use the following command:

jaguar run counterpoise.py filename -keyword igeopt=1 -keyword dftname=x3lyp

We have run binding energy calculations on intermolecular complexes which are held together by a variety of interactions, including hydrogen bonds, π-π interactions, π-cation interactions, halogen bonding, chalcogen bonding, dispersion, and strong electrostatic (ionic) interactions. The most accurate and computationally practical protocol that we have found is M06-2X/cc-pVTZ(-f)+, which usually has a binding energy error of less than 1 kcal/mol when compared to basis set extrapolated CCSD(T) values. One exception is the case of π-cation complexes in which the cation is a metal. Here, no DFT method that we have tried is able to capture the very strong polarizing effect of the metal cation on the π electrons, and the binding energy error is about 5 kcal/mol. For interaction energies of π-cation complexes we recommend using the B3LYP-MM method which contains semiempirical a posteriori corrections that capture effects of such interactions—see B3LYP-MM A Posteriori-Corrected Functional for more detail. However, when the cation is polyatomic, like ammonium, the error drops to about 1 kcal/mol. LMP2 is comparable, and in fact it is no better than M06-2X when treating the dispersively bound complexes in Hobza’s S22 database [266], but it is much more expensive than M06-2X.